Project

Partie 1 : Analyse descriptive

#Mettre le year format date
data_nat_clean$year <- str_pad(data_nat_clean$year,5,"right")
data_nat_clean$year <- str_replace(data_nat_clean$year," ","-31-12")
data_nat_clean$year <- as.Date(data_nat_clean$year, format="%Y-%d-%m")

#Calculer le nombre de prénom unique
distinct_names <- data_nat_clean %>% 
                  group_by(year) %>% 
                  distinct(firstname) %>%
                  summarise(n_names = n())

#Plot le résultat                  
plot_distinct_names <- 
  distinct_names %>% 
  ggplot(aes(x=year, y=n_names)) + 
  geom_line(size=1.2, color="blue") +         
  ggtitle("Number of French unique names from 1900 to 2018") +
  theme(plot.title = element_text(hjust = 0.5)) +
  xlab("Year") + ylab("Number of unique names")

plot_distinct_names

#Ajoute la différence selon l'année
distinct_names$ratio <- ((distinct_names$n_names-lag(distinct_names$n_names))/
                           (distinct_names$n_names+lag(distinct_names$n_names))*100)

#Plot la différence d'augmentation
plot_ratio <- distinct_names %>% 
  ggplot(aes(x=year, y=ratio)) + 
  geom_hline(yintercept=0, size = 1) +
  geom_line(size = 1.2, color = "red") +
  ggtitle("Evolution of names from 1900 to 2018") +
  theme(plot.title = element_text(hjust = 0.5)) +
  xlab("Year") + ylab("Percentage of difference")
plot_ratio

## Warning: Removed 1 rows containing missing values (geom_path).

Working with department data

region <-data_dpt_clean %>% 
  mutate(region=case_when(
  department %in% c("95", "78", "91", "77","94","92","93","75") ~ "Ile-de-France",
  department %in% c("08","51","10","52") ~ "Champagne-Ardenne",
  department %in% c("02","60","80") ~ "Picardie",
  department %in% c("76", "27") ~ "Haute-normandie",
  department %in% c("18","28","36","37", "41", "45") ~ "Centre",
  department %in% c("14","50","61") ~ "Basse-Normandie",
  department %in% c("21", "58", "71", "89") ~ "Bourgogne",
  department %in% c("59","62") ~ "Nord-Pas-de-Calais",
  department %in% c("54","55","57", "88") ~ "Lorraine",
  department %in% c("67", "68") ~ "Alsace",
  department %in% c("25","39","70","90") ~ "Franche-Comté",
  department %in% c("44","49","53", "72","85") ~ "Pays de la Loire",
  department %in% c("22", "29", "35", "56") ~ "Bretagne",
  department %in% c("16","17","79","86") ~ "Poitou-CharenteS",
  department %in% c("24","33","40","47","64") ~ "Aquitaine",
  department %in% c("09", "12", "31", "32","46","65","81","82") ~ "Midi-Pyrenées",
  department %in% c("19","23","87") ~ "Limousin",
  department %in% c("01","07","26","38","42","69","73","74") ~ "Rhone-Alpes",
  department %in% c("03", "15", "43", "63") ~ "Auvergne",
  department %in% c("11","30","34","48", "66") ~ "Languedoc-Roussillon",
  department %in% c("04","05","06","13", "83","84") ~ "Provence-Alpes-Cote d'Azur ",
  department %in% c("20") ~ "Corse",
  department %in% c("971","972","973","974") ~ "Overseas territories",
))

table <- region %>% 
  group_by(year, sex, region, firstname) %>% 
  mutate(number = sum(number)) %>% 
  ungroup() %>% 
  group_by(year, sex, region) %>% 
  select(sex, firstname, year, number, region) %>% 
  unique() %>% 
  filter(number == max(number))

boys <- table %>% filter(sex==1)
girls <- table %>% filter(sex==2)

girls %>% plot_ly(
  x = ~region, 
  y = ~number,
  textposition = 'auto',
  frame = ~year, 
  text = ~firstname, 
  hoverinfo = "number",
  type = 'bar',
  marker = list(color = 'rgb(255,192,203)'))%>%
  layout (
    xaxis = list( title = "French regions"),
    yaxis=list(title="Number of births"),
    title = "Evolution of the most popular girl French firstname according to the region from 1900 to 2018",
    showlegend = FALSE)

boys %>% plot_ly(
  x = ~region, 
  y = ~number,
  textposition = 'auto',
  frame = ~year, 
  text = ~firstname, 
  hoverinfo = "number",
  type = 'bar',
  marker = list(color = 'rgb(135,206,235)'))%>%
  layout (
    xaxis = list( title = "French regions"),
    yaxis=list(title="Number of births"),
    title = "Evolution of the most popular boy French firstname according to the region from 1900 to 2018",
    showlegend = FALSE)